Outcomes of Patients Hospitalized for Severe Acute Alcoholic Hepatitis

Leonard Noronha, MD; Erin FitzGerald, DO; J. Rush Pierce, Jr, MD, MPH


South Med J. 2019;112(7):363-368. 

In This Article

Abstract and Introduction


Objectives: Severe acute alcoholic hepatitis is a serious condition in individuals who consume significant quantities of alcohol. We aimed to identify risk factors for short-term mortality with this illness.

Methods: Patients with severe acute alcoholic hepatitis admitted to our academic medical center from 2010 to 2012 were identified. Demographic features, laboratory values, and patient outcomes were recorded. In-hospital mortality and transfer to inpatient hospice were combined to calculate overall inpatient mortality.

Results: A total of 251 hospitalizations of 191 patients were identified. The average age was 43.1 years (standard deviation 9.55). Most patients were men (80.6%). Compared with all adult patients admitted to internal medicine services during the same period, patients self-reporting Native American and Hispanic race/ethnicity were overrepresented (11.1% vs 34.0% and 14.8% vs 27.7%, χ 2 P < 0.0001). In-hospital mortality was 20.3%. Another 10% of patients were transferred to inpatient hospice facilities. In the multivariate analysis, higher overall inpatient mortality was associated with an admission bilirubin >20 mg/dL (odds ratio 4.59). Of the patients, 11.9% were readmitted with a complication within 30 days—most commonly septic shock. Of the readmitted patients, the overall inpatient mortality was 86.2%.

Conclusions: This study confirms the severity of illness among patients with severe acute alcoholic hepatitis. Patients with the highest total bilirubin levels on admission had the highest overall inpatient mortality. Readmission was a strong predictor of overall in-hospital mortality.


Acute alcoholic hepatitis is a serious condition of young and middle-aged adults who consume alcohol daily. Patients with acute alcoholic hepatitis generally have a long history (typically >10 years) of heavy alcohol consumption. Patients usually present with malaise, muscle weakness, abdominal pain, and jaundice. A small number of these patients report that they discontinued alcohol ingestion a few weeks before onset of the illness. Acute alcoholic hepatitis is distinguished from alcohol-related cirrhosis by acute onset of symptoms and specific laboratory features that include elevation in total bilirubin (typically >5 mg/dL), elevation of hepatic transaminases (<400 U/L), ratio of aspartate aminotransferase to alanine aminotransferase >2, leukocytosis (>10,000 leukocytes per cubic millimeter), and an elevated international normalized ratio. Patients with more severe forms of acute alcoholic hepatitis often have long hospitalizations and a high 90-day mortality rate (40%–90%).[1–7] Those who survive may return to a condition of fatty liver or progress to irreversible cirrhosis. It is common for patients with acute alcoholic hepatitis to have encephalopathy, portal hypertension, esophageal variceal bleeding, and/or ascites.

There are several validated prognostic scoring tools[6,7] that predict 90-day mortality for patients with acute alcoholic hepatitis. Expert opinions from the European Association for the Study of the Liver[6] and the American College of Gastroenterology[7] recommend treatment with prednisolone (American College of Gastroenterology) or either prednisolone or pentoxifylline (European Association for the Study of the Liver) for patients with severe acute alcoholic hepatitis. Severe acute alcoholic hepatitis is defined by a Maddrey discriminant function (DF) >32. A multicenter prospective randomized trial[8] failed to demonstrate the efficacy of these pharmacologic treatments. Nonpharmacologic treatment recommendations include cessation from alcohol and nutritional support.[6,7]

A large number of patients with acute alcoholic hepatitis are admitted to our institution, which features a unique mix of ethnic groups. Our goal was to better understand our own local experience with this disease and to provide more accurate prognostic information to patients and their families. We conducted a descriptive study and performed some limited hypothesis testing about the risk factors associated with mortality.